AN OLD WORKFRAME FOR A NEW PROBLEM: A CLASSICAL MODEL FOR UNSUPERVISED ADAPTIVE FILTERING
Maha Mohamed Elsabrouty · International Conference on Aerospace Sciences and Aviation Technology · 2007
A classical adaptive filtering model of the problem of instantaneous blind signal separation, or what is formally known as unsupervised adaptive filtering is presented. This classical form helps understanding the well-known superior behaviour of the natural gradient solution to the blind separation problem. A new RLS-based algorithm is developed using this classical model. The algorithm provides improved on-line separation speed under the same steady state error compared to the natural gradient algorithm without requiring pre-whitening.